Ontology Enhancement Method and System

1. An ontology enhancement method, comprising: step A: receiving at least an input information request; step B: based on a base ontology, expanding each input information request to produce at least an expanded information request of each corresponding input information request; step C: based on a searching model, according to each expanded information request, searching a file collection to obtain searching results of each corresponding expanded information request; step D: according to each searching result, extracting a plurality of candidate knowledge concepts of each corresponding searching result; and step E: selectively adding the candidate knowledge concepts of each searching result into the base ontology to enhance the base ontology, wherein the base ontology being enhanced is denoted as an enhanced ontology; wherein the step E comprises: step E1: selecting a lowest non-selected one from the expanding information requests; step E2: selecting an i th one of the candidate knowledge concepts of the expanding information request, wherein an initial value of i is 1; step E3: checking whether the selected candidate knowledge concept has been added in the base ontology; step E4: if the selected candidate knowledge concept has not been added in the base ontology, checking whether the selected candidate knowledge concept is the same as the i th one of the candidate knowledge concept of the other expanding information request at the same level; step E5: if the selected candidate knowledge concept is not the same as the i th one of the candidate knowledge concept of the other expanding information request at the same level, adding the selected candidate knowledge concept into the base ontology, and increasing i by 1, i.e. i=i+1; step E6: if the selected candidate knowledge concept is the same as the i th one of the candidate knowledge concept of the other expanding information request at the same level, checking whether a rank of the selected candidate knowledge concept in its father node's expanding information request is larger than a rank of the other same candidate knowledge concept in its father node's expanding information request, wherein a node of the expanding information request of the other same candidate knowledge concept and a node of the expanding information request of the selected candidate knowledge concept are at the same level; step E7: if the rank of the selected candidate knowledge concept in its father node's expanding information request is larger than the rank of the other same candidate knowledge concept in its father node's expanding information request, adding the selected candidate knowledge concept into the base ontology, and increasing i by 1; step E8: if the selected candidate knowledge concept has been added in the base ontology, increasing i by 1; step E9: checking whether i is larger than K, wherein K is a number which the candidate knowledge concepts of the selected expanding information request are in top ranks; step E10: if i is not larger than K, going back to execute the step E2; step E11: if i is larger than K, going back to execute step E2, checking whether any non-selected expanding information request exists; step E12: if at least one non-selected expanding information request exists, going back to execute the step E1; and step E13: if all of the expanding information requests have been selected, ending the step E.

2. The ontology enhancement method according to claim 1 , further comprising: step F: determining whether the enhanced ontology should be modified; and step G: when the enhanced ontology is determined to be modified, modifying the enhanced ontology.

3. The ontology enhancement method according to claim 1 , further comprising: step H: determining whether the enhanced ontology should be further enhanced; and step I: when the enhanced ontology is determined to be further enhanced, repeating step A through step E, and step H.

4. The ontology enhancement method according to claim 2 , wherein an user interface is provided to at least an user to determine whether the enhanced ontology should be modified.

5. The ontology enhancement method according to claim 3 , wherein an user interface is provided to at least an user to determine whether the enhanced ontology should be further enhanced.

6. The ontology enhancement method according to claim 4 , wherein the user is a general user or a specific field expert, and different weighting values are assigned to the general user and the specific field expert.

7. The ontology enhancement method according to claim 5 , wherein the user is a general user or a specific field expert, and different weighting values are assigned to the general user and the specific field expert.

8. The ontology enhancement method according to claim 1 , wherein the searching model is a vector space searching model.

9. The ontology enhancement method according to claim 1 , wherein for each searching result, an n-gram segmentation method is used to extract the candidate knowledge concepts of the corresponding searching result.

10. An ontology enhancement system, comprising: a server, comprising: an information request expanding module, based on a base ontology, expanding each input information request to produce at least an expanded information request of each corresponding input information request; an information searching module, based on a searching model, according to each expanded information request, searching a file collection to obtain searching results of each corresponding expanding information request, and according to each searching result, extracting a plurality of candidate knowledge concepts of each corresponding searching result; and an enhancement module, selectively adding the candidate knowledge concepts of each searching result into the base ontology to enhance the base ontology, wherein the base ontology being enhanced is denoted as an enhanced ontology; wherein selectively adding the candidate knowledge concepts of each searching result into the base ontology comprises: step E1: selecting a lowest non-selected one from the expanding information requests; step E2: selecting an i th one of the candidate knowledge concepts of the expanding information request, wherein an initial value of i is 1; step E3: checking whether the selected candidate knowledge concept has been added in the base ontology; step E4: if the selected candidate knowledge concept has not been added in the base ontology, checking whether the selected candidate knowledge concept is the same as the i th one of the candidate knowledge concept of the other expanding information request at the same level; step E5: if the selected candidate knowledge concept is not the same as the i th one of the candidate knowledge concept of the other expanding information request at the same level, adding the selected candidate knowledge concept into the base ontology, and increasing i by 1, i.e. i=i+1; step E6: if the selected candidate knowledge concept is the same as the i th one of the candidate knowledge concept of the other expanding information request at the same level, checking whether a rank of the selected candidate knowledge concept in its father node's expanding information request is larger than a rank of the other same candidate knowledge concept in its father node's expanding information request, wherein a node of the expanding information request of the other same candidate knowledge concept and a node of the expanding information request of the selected candidate knowledge concept are at the same level; step E7: if the rank of the selected candidate knowledge concept in its father node's expanding information request is larger than the rank of the other same candidate knowledge concept in its father node's expanding information request, adding the selected candidate knowledge concept into the base ontology, and increasing i by 1; step E8: if the selected candidate knowledge concept has been added in the base ontology, increasing i by 1; step E9: checking whether i is larger than K, wherein K is a number which the candidate knowledge concepts of the selected expanding information request are in top ranks; step E10: if i is not larger than K, going back to execute the step E2; step E11: if i is larger than K, going back to execute step E2, checking whether any non-selected expanding information request exists; step E12: if at least one non-selected expanding information request exists, going back to execute the step E1; and step E13: if all of the expanding information requests have been selected, ending selectively adding the candidate knowledge concepts of each searching result into the base ontology.

11. The ontology enhancement system according to claim 10 , further comprising: a database, being included in the server or an external database out of the server, wherein the database stores the ontology and the file collection.

12. The ontology enhancement system according to claim 10 , further comprising: an user terminal, comprising an input module for providing an user to input at least the input information request.

13. The ontology enhancement system according to claim 12 , further comprising: a network, wherein the server and the user terminal are linked to each other through the network.

14. The ontology enhancement system according to claim 10 , wherein the enhancement module determines whether the enhanced ontology should be modified or further enhanced; when the enhanced ontology is determined to be modified, the enhancement module modifies the enhanced ontology; when the enhanced ontology is determined to be further enhanced, the enhancement module further enhances the enhanced ontology.

15. The ontology enhancement system according to claim 14 , wherein an user interface is provided to at least an user to determine whether the enhanced ontology should be modified or further enhanced.

16. The ontology enhancement system according to claim 15 , wherein the user is a general user or a specific field expert, and different weighting values are assigned to the general user and the specific field expert.

17. The ontology enhancement system according to claim 10 , wherein the searching model is a vector space searching model.

18. The ontology enhancement system according to claim 10 , wherein for each searching result, an n-gram segmentation method is used to extract the candidate knowledge concepts of the corresponding searching result.